Multidimensional Adaptive Testing with a Minimum Error-Variance Criterion
- 1 December 1999
- journal article
- Published by American Educational Research Association (AERA) in Journal of Educational and Behavioral Statistics
- Vol. 24 (4) , 398-412
- https://doi.org/10.3102/10769986024004398
Abstract
Adaptive testing under a multidimensional logistic response model is addressed. An algorithm is proposed that minimizes the (asymptotic) variance of the maximum-likelihood estimator of a linear combination of abilities of interest. The criterion results in a closed-form expression that is easy to evaluate. In addition, it is shown how the algorithm can be modified if the interest is in a test with a "simple ability structure". The statistical properties of the adaptive ML estimator are demonstrated for a two-dimensional item pool with several linear combinations of the abilities.Keywords
This publication has 11 references indexed in Scilit:
- A Linear Logistic Multidimensional Model for Dichotomous Item Response DataPublished by Springer Nature ,1997
- Normal-Ogive Multidimensional ModelPublished by Springer Nature ,1997
- Graphical Representation of Multidimensional Item Response Theory AnalysesApplied Psychological Measurement, 1996
- Multidimensional Computerized Adaptive Testing in a Certification or Licensure ContextApplied Psychological Measurement, 1996
- Assembling Tests for the Measurement of Multiple TraitsApplied Psychological Measurement, 1996
- Multidimensional Adaptive TestingPsychometrika, 1996
- Creating a Test Information Profile for a Two-Dimensional Latent SpaceApplied Psychological Measurement, 1994
- The Discriminating Power of Items That Measure More Than One DimensionApplied Psychological Measurement, 1991
- The Difficulty of Test Items That Measure More Than One AbilityApplied Psychological Measurement, 1985
- Normal Ogive Model on the Continuous Response Level in the Multidimensional Latent SpacePsychometrika, 1974